In conjunction with our ongoing effort to develop a new generation of data collection and flow cytometry instrument control hardware (the DiDAC project), a new generation of data display and analysis software is also being developed. (This software is based on the commercially available package known as IDL.) As a result, our data analysis and visualization capabilities will be greatly enhanced. Furthermore, this package will allow us to have, integrated into one application program, all of the analytical and graphical functions that are currently spread across several disparate programs. Many new data display and manipulation capabilities have already been incorporated. Major new features in IDLYK include: The graphical user interface was reorganized to be more consistent with MAC style programming conventions and to facilitate use of the program by naive users. Added the capability to do "Quadrant" statistics (useful for immunophenotyping) as well as "Box" statistics (rectangular regions) on 2D graphs. Many new features were added to control the format of 2D graphs: wire frame surface plots and shaded surface plots, solid filled contour plots, graphs resolution control and smoothing for surface & contour plots independently. Added the capability to have marker regions distinct from Regions of Interest (RIOs); ROIs can be used to both edit the raw data set and highlight events across all of the parameters in a data set. Markers can only set regions on 1D or 2D graphs and integrate events contained in them (Cvs and means are also calculated). The program can now draw 3D isosurfaces from 3D plots (3 parameter histograms). Translational and rotational control is also provided. Neural Net clustering has been added in order to facilitate comprehension of complex, multi-dimensional data sets. The simple competitive learning net implemented seems fairly effective at clustering most flow cytometry data sets. This needs to be characterized in a formal, rigorous manner. The cluster number assignments made by the neural net clustering can be incorporated into the original data file using the analytical data inclusion provisions of the FCS data file format. This information can be extracted for use in subsequent analysis which will facilitate the characterization of the neural net clustering algorithm. Both 1D & 2D histogram file, the FCS format, can be saved from raw listmode data files. This is to facilitate the creation of composite graphs from multiple files as well as to be able to process histogram data in other programs such as "Multicycle" for cell cycle analysis. The capability to create composite graphs of mixed types from multiple data files has been added. The ultimate goal of this provision is to allow the creation of publication quality graphs, directly from the flow cytometry data, which are directly exportable to desktop publishing software packages. A feature similar to BD's "Mousetraks" has been incorporated into IDLYK. This allows the user to 'train' the program to execute a series of operations on a sequence of data files by actually performing the sequence of steps one time. Once the program has been trained, it can execute that sequence any number of times on selected sets of data files. The ability to configure a custom graph has been worked and is nearly complete. This will allow the user to configure exactly the graphs desired from the currently open data file. For example, parameters 2,4, and 6 can be displayed as one dimensional histograms along side two dimensional contour plots of parameters 3 vs. 5 and 2 vs. 6. Once defined, the configuration can be saved for use in subsequent sessions. This program 'DNA Sizer" for analysis of DNA fragment sizing data is complete and in use everyday.